Classification of diabetic retinopathy using unlabeled data and knowledge distillation.
Sajjad AbbasiMohsen HajabdollahiPejman KhadiviNader KarimiRoshanak RoshandelShahram ShiraniShadrokh SamaviPublished in: Artif. Intell. Medicine (2021)
Keyphrases
- unlabeled data
- labeled data
- semi supervised learning
- co training
- supervised learning
- semi supervised classification
- class labels
- labeled and unlabeled data
- semi supervised
- text classification
- improve the classification accuracy
- supervised learning algorithms
- training set
- active learning
- label information
- prior knowledge
- classification accuracy
- learning algorithm
- training data
- decision boundary
- text categorization
- labeled examples
- semi supervised learning algorithms
- domain knowledge
- classification algorithm
- unsupervised learning
- labeled training data
- labeled instances
- transfer learning
- data sets
- supervised and semi supervised
- data points
- knowledge representation
- training examples
- diabetic retinopathy
- feature extraction
- knowledge discovery
- knowledge based systems
- feature space
- machine learning
- retinal images
- model selection
- naive bayes
- pairwise
- expert systems
- automatic detection
- domain experts
- information retrieval